CN112467686B - Power distribution network fault recovery method, device, control equipment and storage medium - Google Patents
Power distribution network fault recovery method, device, control equipment and storage medium Download PDFInfo
- Publication number
- CN112467686B CN112467686B CN202011214337.4A CN202011214337A CN112467686B CN 112467686 B CN112467686 B CN 112467686B CN 202011214337 A CN202011214337 A CN 202011214337A CN 112467686 B CN112467686 B CN 112467686B
- Authority
- CN
- China
- Prior art keywords
- distribution network
- power distribution
- control
- fault
- control action
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H3/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection
- H02H3/02—Details
- H02H3/06—Details with automatic reconnection
- H02H3/063—Details concerning the co-operation of many similar arrangements, e.g. in a network
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Abstract
The application relates to a power distribution network fault recovery method, a device, control equipment and a storage medium. In the whole fault recovery process, a model is not required to be built, a large amount of calculation is not required, only the historical self-healing action database is required to be queried, and after simulated fault recovery, an optimal control action is selected according to a plurality of fault removal overheads to carry out fault recovery on the power distribution network. The method solves the technical problems that in the prior art, the calculated amount is large and the complexity is high in the solving process, and the optimal solution is difficult to find, and achieves the technical effect of greatly reducing the calculated amount and the calculated complexity on the premise of ensuring that the optimal solution can be found.
Description
Technical Field
The present application relates to the field of power grid security technologies, and in particular, to a power distribution network fault recovery method, a device, a control device, and a storage medium.
Background
With the proposal of the concept of the power distribution network, the focus of power grid safety attention extends from small-scale and frequent faults to large-scale and disastrous events, recovery control also gradually covers the relatively fragile power distribution network from a power transmission network, and for the recovery control of the power distribution network, a new idea for coping with extreme natural disasters is provided from the aspects of defense and more rapid recovery after disaster.
The current method for quickly recovering the power distribution network after disaster is to respectively establish an adaptive multi-stage fault recovery model aiming at different fault stages of the power distribution network under extreme disasters, and perform optimization solution by adopting a traditional optimization planning method (such as a commercial optimization solver) so as to obtain an optimal recovery control scheme. However, the conventional optimization planning method only considers very limited typical working conditions and scenes, and under the extreme external environment or in the actual fault recovery process, the environment is complex, the fault types are numerous and the action linkage effect is serious, so that the calculation amount in the solving process is large and the complexity degree is high, and therefore, the optimal solution is difficult to find.
Disclosure of Invention
Based on the foregoing, it is necessary to provide a power distribution network fault recovery method, a device, a control apparatus and a storage medium for solving the above technical problems.
In a first aspect, a power distribution network fault recovery method is provided, and the method includes:
under the condition that the power distribution network fails, acquiring the current observable state of the power distribution network, wherein the observable state comprises a failure state parameter and an environmental parameter causing the power distribution network to fail;
inquiring a historical self-healing action database according to the current observable state to obtain a control action set corresponding to the current observable state, wherein a plurality of corresponding relations between the observable state and the control action are stored in the historical self-healing action database;
sequentially simulating and executing control actions in the control action set, and acquiring fault removal overhead corresponding to the executed control actions after each execution;
determining an optimal control action from the control action set according to the obtained multiple fault removal overheads;
and performing fault recovery on the power distribution network based on the optimal control action.
In an alternative embodiment of the present application, after each execution, obtaining a troubleshooting overhead corresponding to the executed control action includes: after each execution, determining the execution cost corresponding to the executed control action and the compensation value of the flexible load in the power distribution network to the execution cost; after each execution, determining the fault removal overhead according to the execution overhead corresponding to the executed control action and the compensation value.
In an alternative embodiment of the present application, determining the troubleshooting overhead according to the execution overhead and the compensation value corresponding to the executed control action includes: and determining a first difference value between the execution overhead and the compensation value to obtain the fault removal overhead.
In an alternative embodiment of the present application, determining an optimal control action from a set of control actions based on the resulting plurality of troubleshooting overheads comprises: respectively calculating second difference values between the plurality of fault clearing overheads and the preset fault clearing overheads to obtain a plurality of second difference values; and determining the control action corresponding to the minimum numerical value in the second difference values in the control action set as the optimal control action.
In an alternative embodiment of the application, the observable states include distributed mains power, flexible load power and line damage status in the distribution network.
In an alternative embodiment of the present application, further comprising: and carrying out normalization processing on the distributed power supply power, the flexible load power and the line damaged state to obtain an observable state.
In an alternative embodiment of the application, the set of control actions includes the action state of tie switches in the distribution network.
In a second aspect, there is provided a fault recovery apparatus for a power distribution network, the apparatus comprising: the system comprises a state acquisition module, a control action determination module, an overhead determination module, an optimal action determination module and a fault recovery module.
The state acquisition module is used for acquiring the current observable state of the power distribution network under the condition that the power distribution network fails, wherein the observable state comprises a failure state parameter and an environmental parameter which causes the power distribution network to fail;
the control action determining module is used for inquiring a historical self-healing action database according to the current observable state to obtain a control action set corresponding to the current observable state, and a plurality of corresponding relations between the observable state and the control actions are stored in the historical self-healing action database;
the overhead determining module is used for sequentially simulating and executing control actions in the control action set, and acquiring fault elimination overhead corresponding to the executed control actions after each execution;
the optimal action determining module is used for determining an optimal control action from the control action set according to the obtained multiple fault removal overheads;
the fault recovery module is used for carrying out fault recovery on the power distribution network based on the optimal control action.
In an alternative embodiment of the application, the overhead determination module comprises: a first overhead module and a second overhead module. The first overhead module is used for determining the execution overhead corresponding to the executed control action and the compensation value of the flexible load in the power distribution network on the execution overhead after each execution; the second overhead module is used for determining fault removal overhead according to the execution overhead and the compensation value corresponding to the executed control action after each execution.
In an alternative embodiment of the present application, the second overhead module is specifically configured to determine a first difference between the execution overhead and the compensation value, and obtain the troubleshooting overhead.
In an alternative embodiment of the present application, the optimal action determination module includes: the first motion determination sub-module and the second motion determination sub-module. The first action determining submodule is used for respectively calculating second difference values between the plurality of fault removal overheads and preset fault removal overheads to obtain a plurality of second difference values; the second action determining submodule is used for determining that the control action corresponding to the minimum numerical value in the second difference values in the control action set is the optimal control action.
In an alternative embodiment of the application, the observable states include distributed mains power, flexible load power and line damage status in the distribution network.
In an optional embodiment of the application, the system further comprises a processing module, wherein the processing module is used for normalizing the distributed power supply, the flexible load power and the line damage state to obtain an observable state.
In an alternative embodiment of the application, the set of control actions includes the action state of tie switches in the distribution network.
In a third aspect, there is provided a control device comprising a memory storing a computer program and a processor implementing the steps of the method as above when the processor executes the computer program.
In a fourth aspect, a computer readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method as above.
The embodiment of the application provides a power distribution network fault recovery method, which is used for acquiring the observable state of a power distribution network in real time under the condition that the power distribution network fails, and inquiring a historical self-healing action database according to the current observable state to obtain a control action set. And performing control in the control action set in sequence in a simulation mode, determining to obtain an optimal control action through fault removal overhead after each control action is performed, and finally performing fault recovery on the power distribution network based on the optimal control action. In the whole fault recovery process, a model is not required to be built, a large amount of calculation is not required, only the historical self-healing action database is required to be queried, and after simulated fault recovery, an optimal control action is selected according to a plurality of fault removal overheads to carry out fault recovery on the power distribution network. The power distribution network fault recovery method provided by the embodiment of the application solves the technical problems that in the prior art, the calculated amount is large and the complexity is high in the solving process, and the optimal solution is difficult to find, and achieves the technical effect of greatly reducing the calculated amount and the calculated complexity on the premise of ensuring that the optimal solution can be found.
Drawings
FIG. 1 is an application environment diagram of a power distribution network fault recovery method in one embodiment;
FIG. 2 is a flow chart of a method for recovering from a power distribution network fault in one embodiment;
FIG. 3 is a flow chart of a method for recovering from a power distribution network fault in one embodiment;
FIG. 4 is a flow chart of a method for recovering from a power distribution network fault in one embodiment;
FIG. 5 is a flow chart of a method for recovering from a power distribution network fault in one embodiment;
FIG. 6 is a flow chart of a method of power distribution network fault recovery in one embodiment;
FIG. 7 is a block diagram of a power distribution network fault recovery apparatus in one embodiment;
fig. 8 is a block diagram of a control device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
At present, a traditional optimization planning method such as a commercial optimization solving method is generally adopted for post-disaster recovery of a power distribution network to control recovery faults. However, in extreme disasters, the fault derivation may go through different stages, and different fault recovery models need to be continuously built or updated to adapt to the fault recovery of the different stages. The fault recovery model is complex to build and takes long time, so that the current fault recovery model only models very limited typical working conditions and scenes, but in an extreme external environment or in an actual fault recovery process, because the environment is complex, the fault types are numerous and the action linkage effect is serious, the calculation amount in the solving process is large and the complexity degree is high, and therefore, an optimal solution is difficult to find.
In view of the above, the embodiment of the application provides a power distribution network fault recovery method, which is characterized in that when a power distribution network fails, observable states of the power distribution network are obtained in real time, a historical self-healing action database is queried according to the current observable states, and a control action set is obtained. And performing control in the control action set in sequence in a simulation mode, determining to obtain an optimal control action through fault removal overhead after each control action is performed, and finally performing fault recovery on the power distribution network based on the optimal control action. In the whole fault recovery process, a model is not required to be built, a large amount of calculation is not required, only the historical self-healing action database is required to be queried, and after simulated fault recovery, an optimal control action is selected according to a plurality of fault removal overheads to carry out fault recovery on the power distribution network. The power distribution network fault recovery method provided by the embodiment of the application solves the technical problems that in the prior art, the calculated amount is large and the complexity is high in the solving process, and the optimal solution is difficult to find, and achieves the technical effect of greatly reducing the calculated amount and the calculated complexity on the premise of ensuring that the optimal solution can be found.
Next, an implementation environment related to the fault recovery method of the power distribution network provided by the embodiment of the present application will be briefly described.
Referring to fig. 1, the power distribution network fault recovery method provided by the embodiment of the application is applied to a power distribution network, and the power distribution network comprises a power network body, monitoring equipment, control equipment and a control terminal. The monitoring equipment is arranged at different positions of the power grid body and is used for detecting operation signals and environment signals of different nodes or equipment in the power grid body, the control equipment is in signal connection with different electric equipment in the power grid body and is used for controlling the different electric equipment to work, the control terminal is respectively in signal connection with the monitoring equipment and the control equipment and is used for sending control instructions to the control equipment according to the operation signals and the environment signals collected by the monitoring equipment and controlling the action of the control equipment according to the operation signals and the environment signals. The monitoring device may include a voltage acquisition device, a current acquisition device, a resistance meter, a temperature acquisition device, a humidity measurement device, a weather meter, and the like, and the embodiment of the application is not particularly limited.
Referring to fig. 2, an embodiment of the present application provides a fault recovery method for a power distribution network, which may be applied to the power distribution network, where the following embodiment is applied to the control device in fig. 1, and is used to specifically describe fault recovery of the power distribution network, and includes the following steps 201 to 205:
Step 201, under the condition that a power distribution network fails, a control terminal obtains the current observable state of the power distribution network, wherein the observable state comprises a failure state parameter and an environmental parameter causing the power distribution network to fail.
The monitoring equipment is used for collecting parameters such as voltage, current, resistance, switch on-off and the like of the power distribution network in real time, so that the fault state parameters are obtained. The temperature, the humidity, the illumination and the like in the power distribution network are collected in real time through the monitoring equipment, so that the environment parameters which cause the power distribution network to fail are obtained. It should be noted that the fault state parameters are not limited to voltage, current, resistance, and on-off parameters of the switch, but may include any other parameters that can characterize the fault state of the power distribution network. Similarly, the environmental parameters are not limited to temperature, humidity and illumination parameters, but can also include any other parameters that can characterize the environment that causes the power distribution network to fail.
Step 202, the control terminal queries a historical self-healing action database according to the current observable state to obtain a control action set corresponding to the current observable state, and a plurality of corresponding relations between the observable state and the control actions are stored in the historical self-healing action database.
The historical self-healing action database is prestored in the control terminal and can also be stored in other storage equipment of the power distribution network, the control terminal can access the historical self-healing action database in real time, and when the control terminal accesses the historical self-healing action database, all corresponding relations can be queried. The historical self-healing action database comprises a plurality of corresponding relations, and in each corresponding relation, one or more control actions corresponding to one observable state are carried out. Firstly, the control terminal performs information matching in the historical self-healing action database, finds the observable state identical to the current observable state, and then determines the control action historically executed in the observable state according to the corresponding relation. In the recovery from a history failure, a plurality of different control actions may occur in the same observable state, and the plurality of different control actions constitute the control action set described above.
Step 203, the control terminal sequentially simulates and executes the control actions in the control action set, and obtains the fault removal overhead corresponding to the executed control actions after each execution.
The control terminal inputs the control actions in the control action set to the power grid fault model in sequence, the control actions act in the power grid fault model, and the state of the power grid fault model changes. At this time, the control terminal determines the current fault removal cost according to the current power grid fault model recovery condition and other factors such as the electricity consumption during fault recovery, where the fault removal cost refers to the electricity consumption cost or other costs required for removing the power grid fault. For example, when the troubleshooting overhead is a power cost, it may be determined by determining the power required to perform the control action when the fault occurs.
And 204, the control terminal determines the optimal control action from the control action set according to the obtained multiple fault removal overheads.
The control terminal obtains a fault removal cost for each execution of the control action on the power grid fault model, and obtains a plurality of fault removal costs after respectively executing a plurality of control actions. The control terminal determines a minimum cost according to the multiple fault removal costs, and then inquires a control action corresponding to the minimum cost in the control action set to determine that the control action is an optimal control action.
And 205, the control terminal performs fault recovery on the power distribution network based on optimal control braking.
After determining the optimal control action, the control terminal performs control adjustment on the power distribution network body through control equipment, for example, adjusts the running state of the power distribution network body through adjusting certain node switches in the control equipment or restarting certain controllers and the like, so that the faults of the power distribution network are recovered to be normal, the purposes of eliminating the faults in the power distribution network and recovering the faults of the power distribution network are achieved.
The embodiment of the application provides a power distribution network fault recovery method, which is used for acquiring the observable state of a power distribution network in real time under the condition that the power distribution network fails, and inquiring a historical self-healing action database according to the current observable state to obtain a control action set. And performing control in the control action set in sequence in a simulation mode, determining to obtain an optimal control action through fault removal overhead after each control action is performed, and finally performing fault recovery on the power distribution network based on the optimal control action. In the whole fault recovery process, a model is not required to be built, a large amount of calculation is not required, only the historical self-healing action database is required to be queried, and after simulated fault recovery, an optimal control action is selected according to a plurality of fault removal overheads to carry out fault recovery on the power distribution network. The power distribution network fault recovery method provided by the embodiment of the application solves the technical problems that in the prior art, the calculated amount is large and the complexity is high in the solving process, and the optimal solution is difficult to find, and achieves the technical effect of greatly reducing the calculated amount and the calculated complexity on the premise of ensuring that the optimal solution can be found.
In an alternative embodiment of the application, the observable state includes distributed mains power, flexible load power and line damage status in the distribution network.
According to the first aspect, the control terminal acquires the current and the voltage of the distributed power supply and the flexible load in the power distribution network in real time through the current acquisition equipment and the voltage acquisition equipment in the monitoring equipment, and then the power of the distributed power supply and the power of the flexible load are respectively calculated through a power calculation formula. The flexible load is an interruptible load in the distribution network. Alternatively, the observable state may be determined by the following formula:
wherein in formula (1), S t Representing the observable state of the distribution network at time t,representing the output power of the ith distributed power supply at time t, P t j Indicating the power of the flexible load at time t, +.>Indicating the damaged state of the first line at the time t, N DG Representing a set of distributed power supplies, N L Representing a collection of flexible loads, N f Representing a collection of vulnerable lines.
In a second aspect, the control terminal detects whether a line in the power distribution network is damaged through an infrared detector and the like in the monitoring device, and in another possible case, the control terminal compares current data and voltage data acquired in real time through the current acquisition device and the voltage acquisition device with a preset normal working current range and a preset normal working voltage range to determine whether the current line is damaged. Wherein the damaged state includes both damaged and undamaged states.
In an alternative embodiment of the application, the set of control actions includes the action state of tie switches in the distribution network.
The tie switch is an important control device in the power distribution network and is used for controlling whether different transmission lines or electric devices work or not and electric connection between the transmission lines or the electric devices. Meanwhile, the interconnection switch is a fragile and easy-to-damage node in the power distribution network, so that the interconnection switch is used as a control point for fault recovery of the power distribution network, the coverage is wider, and the recovery effect is better. The state of the tie switch is only two types of closing and opening, and can be determined by the following action function, for example:
wherein a is t The control action is indicated to be performed,in order to make the state variable of the mth interconnecting switch at the time t+1, only 0 or 1 has two values, wherein the value 0 represents closing, the value 1 represents opening and N con The method is used for collecting contact switch branches in the power distribution network.
In an alternative embodiment of the present application, further comprising: and the control terminal normalizes the distributed power supply power, the flexible load power and the line damage state to obtain an observable state.
In a first aspect, taking the power of the flexible load as an example, it can be determined by the following formula:
wherein the power P of the flexible load t j The maximum value and the minimum value of the self-healing action can be obtained from a historical self-healing action database or can be obtained according to inquiry in a historical power grid data record.
In the second aspect, if the line fails to break at time t for the line damaged stateThe value is 1, otherwise defaulting to 0. From this, the values of all states after normalization are all 0,1]And the calculated amount is greatly reduced, and the failure recovery efficiency of the power distribution network is improved.
Referring to fig. 3, in an alternative embodiment of the present application, step 203 includes steps 301-302:
step 301, after each execution, the control terminal determines the execution overhead corresponding to the executed control action and the compensation value of the flexible load in the power distribution network to the execution overhead.
After the control terminal executes the control action on the power grid fault model each time, the control terminal calculates and determines the execution overhead in the process of executing the control action and the compensation value of the flexible load in the power distribution network. The execution overhead refers to the amount of power consumed by the power distribution network when performing control actions on the grid fault model. The flexible load refers to an interruptible load in a power grid, namely the flexible load can perform power failure processing in actual fault recovery without consuming electric quantity. In this embodiment, the simulation result of the power grid fault model is combined with actual factors to determine the fault removal overhead, so as to improve the reliability of the power distribution network fault recovery method in the embodiment of the application when fault recovery is actually performed.
And 302, after each execution, the control terminal determines the fault removal overhead according to the execution overhead and the compensation value corresponding to the executed control action.
The control terminal may determine a first difference between the execution overhead and the compensation value, resulting in a troubleshooting overhead. The troubleshooting overhead may be determined by the following formula:
wherein r is t+1 Represents the overhead of the troubleshooting,cost per output power for distributed power supply, +.>Represents the output power of the ith distributed power supply at time t,/>Cost per power for jth flexible load, P t j Representing the power of the flexible load at time t, N DG Representing a set of distributed power supplies, N IL Is a flexible load set in the system.
Optionally, in the embodiment of the present application, the fault removal overhead may also be calculated by using an electricity generation ratio method based on the power failure loss characteristic of the flexible load:
wherein lambda is k The power failure loss of unit electricity quantity when the power failure happens to the flexible load of the node k of the power distribution network, L represents the electricity generation ratio, beta k And (5) weighting the flexible load of the power distribution network node k.
Referring to fig. 4, in an alternative embodiment of the present application, step 204 includes steps 401-402:
in step 401, the control terminal calculates second difference values between the multiple fault clearing overheads and the preset fault clearing overheads, so as to obtain multiple second difference values.
The preset fault removal cost is stored in the control terminal in advance, and when the control terminal obtains one fault removal cost each time, the fault removal cost is compared with the preset fault removal cost stored in the control terminal, and a difference value is calculated to obtain a second difference value. After the control terminal performs multiple comparisons, multiple second difference values can be obtained. The preset troubleshooting overhead may be determined according to historical experience or comprehensive consideration of various factors, and the embodiment is not particularly limited.
Step 402, the control terminal determines that the control action corresponding to the minimum value in the plurality of second difference values in the control action set is the optimal control action.
The control terminal calculates a plurality of second difference values, wherein the second difference values are used for representing the difference value between the corresponding fault clearing expense and the preset fault clearing expense when each control action is executed, the smaller the second difference values are, the closer the fault clearing expense is to the preset fault clearing expense, and particularly when the second difference values are negative values, the fault clearing expense in actual fault restoration is far lower than the preset fault clearing expense, and the control action is better.
Referring to fig. 5, in an alternative embodiment of the present application, the control terminal may further perform network training through the DDQN algorithm to obtain the above-mentioned optimal control action:
And 501, the control terminal determines an action estimation function according to the fault state of the power distribution network and the environmental parameters which cause the power distribution network to fail.
The control terminal fits the fault state estimation function V(s) under the observed state by using two state estimation function networks of the neural network fitting observable state respectively t ) And a dominance estimation function a (s t ,a t ) Wherein the fault state estimation function V (s t ) Is a function of the fault state parameters of the power distribution network, and the dominance estimation function A (s t ,a t ) Is a function of the above-mentioned environmental parameters which lead to a failure of the distribution network, according to which a function V (s t ) And the dominance estimation function A (s t ,a t ) Obtaining an action estimation function:
wherein Q (S) t ,a t ) Action evaluation function, A represents control action set, |A| represents the number of control actions in control action set, V(s) t ) Representing a fault state estimation function, A (s t ,a t ) Representing the dominance estimation function.
In this embodiment, only one optimal control action can be selected in each observable state, only one Q value can be obtained, and the Q value cannot be disassembled into the unique state estimation function V value and the action dominance function a value, so that the action dominance function is set to be an independent action dominance function minus the average value of all action dominance functions in the current state, so as to remove redundant degrees of freedom and improve algorithm stability.
Step 502, the control terminal determines an optimal motion estimation function according to the motion estimation function.
Optionally, the control terminal obtains the target value of the optimal action estimation function by using a bellman equation:
according to the current observable state, updating the action estimation function, wherein the specific formula is as follows:
wherein, in the formulas (7) - (8), the discount factor lambda epsilon [0,1], the learning rate 0< alpha is less than or equal to 1.
Referring to fig. 6, in an alternative embodiment of the present application, step 502 includes steps 601-603:
step 601, the control terminal introduces an epsilon-greedy strategy to select the control actions in the control action set:
wherein epsilon is a fixed constant in an epsilon-greedy strategy, T is the total training times, and T is the current training times.
Step 602, the control terminal screens observable state samples in the historical self-healing action database according to a preset model.
To eliminate timing dependencies between samples over a short period, a memory playback is employed to store a historical self-healing action database. And (3) establishing an experience pool with the capacity of N, storing a sample of the corresponding relation between the observable state of the power distribution network and the control action in each training period, randomly extracting small batches of samples from the experience pool when the number of the samples exceeds the playback starting capacity M, carrying out artificial neural network training, and training the neural network by randomly extracting the samples to avoid the phenomena of overfitting and the like. If the number of samples exceeds the maximum capacity of the experience pool, the earliest sample is removed and then stored in a new sample, so that the neural network is ensured to learn the latest observation state.
And 603, the control terminal determines a loss function according to the action estimation function of each control action.
The DDQN performs forward computation to obtain Q values of all control actions, in this embodiment, the loss function refers to the mean square error between the target Q value and the predicted Q value output by the neural network, and the neural network loss function is determined by formulas (7) - (9), where the specific formula is
Training the neural network by using a small-batch gradient descent method, then acquiring a real-time observable state of the power distribution network by using monitoring equipment, and selecting an estimated maximum action a by using the trained neural network k I.e. the control optimization strategy, i.e. the above-mentioned optimal control actions.
It should be understood that, although the steps in the flowchart are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least a portion of the steps in the figures may include steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a portion of the steps or stages in other steps.
Referring to fig. 7, an embodiment of the present application provides a power distribution network fault recovery apparatus 10, including: a state acquisition module 100, a control action determination module 200, an overhead determination module 300, an optimal action determination module 400, and a fault recovery module 500.
The state acquisition module 100 is configured to acquire, in the event of a power distribution network failure, a current observable state of the power distribution network, where the observable state includes a failure state parameter and an environmental parameter that causes the power distribution network to fail;
the control action determining module 200 is configured to query a historical self-healing action database according to the current observable state, to obtain a control action set corresponding to the current observable state, where the historical self-healing action database stores a plurality of corresponding relations between the observable state and the control action;
the overhead determining module 300 is configured to sequentially simulate and execute control actions in the control action set, and obtain troubleshooting overhead corresponding to the executed control actions after each execution;
the optimal action determining module 400 is configured to determine an optimal control action from the control action set according to the obtained multiple troubleshooting overheads;
The fault recovery module 500 is configured to perform fault recovery on the power distribution network based on the optimal control brake.
In an alternative embodiment of the application, the overhead determination module 300 includes: a first overhead module and a second overhead module.
The first overhead module is used for determining the execution overhead corresponding to the executed control action and the compensation value of the flexible load in the power distribution network for the execution overhead after each execution;
the second overhead module is used for determining the fault elimination overhead according to the execution overhead corresponding to the executed control action and the compensation value after each execution.
In an alternative embodiment of the present application, the second overhead module is specifically configured to determine a first difference between the execution overhead and the compensation value, and obtain the troubleshooting overhead.
In an alternative embodiment of the present application, the optimal action determination module 400 includes: the first motion determination sub-module and the second motion determination sub-module.
The first action determining submodule is used for respectively calculating a plurality of second difference values between the fault clearing expense and preset fault clearing expense to obtain a plurality of second difference values;
the second action determining submodule is used for determining control actions corresponding to minimum numerical values in the second differences in the control action set as the optimal control actions.
In an alternative embodiment of the application, the observable state includes distributed mains power, flexible load power and line damaged status in the distribution network.
In an optional embodiment of the present application, the system further includes a processing module, where the processing module is configured to normalize the distributed power supply, the flexible load power, and the line damaged state to obtain the observable state.
In an alternative embodiment of the application, the set of control actions includes an action state of a tie switch in the distribution network.
The specific limitation of the power distribution network fault recovery apparatus 10 may be referred to as the limitation of the power distribution network fault recovery method hereinabove, and will not be described herein. The various modules in the power distribution network fault recovery apparatus 10 described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the control device, or may be stored in software in a memory in the control device, so that the processor may call and execute operations corresponding to the above modules.
Fig. 8 is a schematic diagram illustrating an internal structure of a control device according to an embodiment of the present application, where the control device may be a server. As shown in fig. 8, the control device includes a processor, a memory, and a communication component connected by a system bus. Wherein the processor is configured to provide computing and control capabilities to support the operation of the overall control device. The memory may include a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The computer program is executable by a processor for implementing a power distribution network fault recovery method provided by the above embodiments. The internal memory provides a cached operating environment for the operating system and computer programs in the non-volatile storage media. The control device may communicate with other control devices (e.g., STAs) through a communication component.
It will be appreciated by those skilled in the art that the structure shown in fig. 8 is merely a block diagram of a portion of the structure associated with the present inventive arrangements and is not limiting of the control device to which the present inventive arrangements are applied, and that a particular control device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, there is provided a control apparatus including: the device comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the following steps when executing the computer program:
under the condition that a power distribution network fails, acquiring the current observable state of the power distribution network, wherein the observable state comprises a failure state parameter and an environmental parameter which causes the power distribution network to fail;
inquiring a historical self-healing action database according to the current observable state to obtain a control action set corresponding to the current observable state, wherein a plurality of corresponding relations between the observable state and the control action are stored in the historical self-healing action database;
sequentially simulating and executing the control actions in the control action set, and acquiring fault removal overhead corresponding to the executed control actions after each execution;
Determining an optimal control action from the control action set according to the obtained multiple fault removal overheads;
and carrying out fault recovery on the power distribution network based on the optimal control brake.
In one embodiment of the application, the processor when executing the computer program further performs the steps of: after each execution, determining the execution cost corresponding to the executed control action and the compensation value of the flexible load in the power distribution network to the execution cost; and after each execution, determining the fault elimination overhead according to the execution overhead corresponding to the executed control action and the compensation value.
In one embodiment of the application, the processor when executing the computer program further performs the steps of: and determining a first difference value between the execution overhead and the compensation value to obtain the fault removal overhead.
In one embodiment of the application, the processor when executing the computer program further performs the steps of: respectively calculating a plurality of second difference values between the fault clearing expense and preset fault clearing expense to obtain a plurality of second difference values; and determining the control brake corresponding to the minimum numerical value in the second differences in the control action set as the optimal control action.
In one embodiment of the application, the observable state includes distributed mains power, flexible load power and line damage status in the distribution network.
In one embodiment of the application, the processor when executing the computer program further performs the steps of: and normalizing the distributed power supply power, the flexible load power and the line damage state to obtain the observable state.
In one embodiment of the application, the set of control actions includes an action state of a tie switch in the power distribution network.
The control device provided in the embodiment of the present application has similar implementation principles and technical effects to those of the above method embodiment, and will not be described herein.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
under the condition that a power distribution network fails, acquiring the current observable state of the power distribution network, wherein the observable state comprises a failure state parameter and an environmental parameter which causes the power distribution network to fail;
inquiring a historical self-healing action database according to the current observable state to obtain a control action set corresponding to the current observable state, wherein a plurality of corresponding relations between the observable state and the control action are stored in the historical self-healing action database;
Sequentially simulating and executing the control actions in the control action set, and acquiring fault removal overhead corresponding to the executed control actions after each execution;
determining an optimal control action from the control action set according to the obtained multiple fault removal overheads;
and carrying out fault recovery on the power distribution network based on the optimal control brake.
In one embodiment of the application, the computer program when executed by the processor further implements the steps of: after each execution, determining the execution cost corresponding to the executed control action and the compensation value of the flexible load in the power distribution network to the execution cost; and after each execution, determining the fault elimination overhead according to the execution overhead corresponding to the executed control action and the compensation value.
In one embodiment of the application, the computer program when executed by the processor further implements the steps of: and determining a first difference value between the execution overhead and the compensation value to obtain the fault removal overhead.
In one embodiment of the application, the computer program when executed by the processor further implements the steps of: respectively calculating a plurality of second difference values between the fault clearing expense and preset fault clearing expense to obtain a plurality of second difference values; and determining the control brake corresponding to the minimum numerical value in the second differences in the control action set as the optimal control action.
In one embodiment of the application, the observable state includes distributed mains power, flexible load power and line damage status in the distribution network.
In one embodiment of the application, the computer program when executed by the processor further implements the steps of: and normalizing the distributed power supply power, the flexible load power and the line damage state to obtain the observable state.
In one embodiment of the application, the set of control actions includes an action state of a tie switch in the power distribution network.
The computer readable storage medium provided in this embodiment has similar principles and technical effects to those of the above method embodiment, and will not be described herein.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in M forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SyMchlimk) DRAM (SLDRAM), memory bus (RaMbus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. A method for recovering from a power distribution network fault, the method comprising:
under the condition that a power distribution network fails, acquiring the current observable state of the power distribution network, wherein the observable state comprises a failure state parameter and an environmental parameter which causes the power distribution network to fail;
inquiring a historical self-healing action database according to the current observable state to obtain a control action set corresponding to the current observable state, wherein a plurality of corresponding relations between the observable state and the control action are stored in the historical self-healing action database;
Sequentially simulating and executing the control actions in the control action set, and acquiring fault removal overhead corresponding to the executed control actions after each execution;
after each execution, determining the execution cost corresponding to the executed control action and the compensation value of the flexible load in the power distribution network to the execution cost;
after each execution, determining the fault elimination overhead according to the execution overhead and the compensation value corresponding to the executed control action;
determining an optimal control action from the control action set according to the obtained multiple fault removal overheads;
and carrying out fault recovery on the power distribution network based on the optimal control brake.
2. The power distribution network fault recovery method according to claim 1, wherein the determining the fault removal overhead from the execution overhead and the compensation value corresponding to the executed control action includes:
and determining a first difference value between the execution overhead and the compensation value to obtain the fault removal overhead.
3. The power distribution network fault recovery method according to claim 1, wherein said determining an optimal control action from the set of control actions based on the resulting plurality of troubleshooting overheads comprises:
Respectively calculating a plurality of second difference values between the fault clearing expense and preset fault clearing expense to obtain a plurality of second difference values;
and determining the control brake corresponding to the minimum numerical value in the second differences in the control action set as the optimal control action.
4. The power distribution network fault recovery method of claim 1, wherein the observable states include distributed power supply power, flexible load power, and line damaged states in the power distribution network.
5. The power distribution network fault recovery method of claim 4, further comprising:
and normalizing the distributed power supply power, the flexible load power and the line damage state to obtain the observable state.
6. The power distribution network fault recovery method of claim 1, wherein the control action set comprises an action state of a tie switch in the power distribution network.
7. The power distribution network fault recovery method according to claim 1, wherein the fault removal overhead is calculated by an electricity generation ratio method based on flexible load power outage loss characteristics.
8. A power distribution network fault recovery apparatus, the apparatus comprising:
The system comprises a state acquisition module, a state detection module and a state control module, wherein the state acquisition module is used for acquiring the current observable state of the power distribution network under the condition that the power distribution network fails, and the observable state comprises a failure state parameter and an environment parameter which causes the power distribution network to fail;
the control action determining module is used for inquiring a historical self-healing action database according to the current observable state to obtain a control action set corresponding to the current observable state, and a plurality of corresponding relations between the observable state and the control action are stored in the historical self-healing action database;
the overhead determining module is used for sequentially simulating and executing the control actions in the control action set, and acquiring fault elimination overhead corresponding to the executed control actions after each execution;
the optimal action determining module is used for determining optimal control actions from the control action set according to the obtained multiple fault removal overheads;
and the fault recovery module is used for carrying out fault recovery on the power distribution network based on the optimal control brake.
9. A control device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011214337.4A CN112467686B (en) | 2020-11-04 | 2020-11-04 | Power distribution network fault recovery method, device, control equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011214337.4A CN112467686B (en) | 2020-11-04 | 2020-11-04 | Power distribution network fault recovery method, device, control equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112467686A CN112467686A (en) | 2021-03-09 |
| CN112467686B true CN112467686B (en) | 2023-09-01 |
Family
ID=74835942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011214337.4A Active CN112467686B (en) | 2020-11-04 | 2020-11-04 | Power distribution network fault recovery method, device, control equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112467686B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113541092B (en) * | 2021-09-15 | 2021-11-26 | 国网天津市电力公司城南供电分公司 | Power distribution network relay protection artificial intelligence setting method based on reinforcement learning algorithm |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018025968A (en) * | 2016-08-10 | 2018-02-15 | 日本電信電話株式会社 | Restoration control system and method |
| CN108123439A (en) * | 2018-01-03 | 2018-06-05 | 燕山大学 | A kind of repairing of active distribution network multiple faults and restoration methods based on multi-agent system |
| CN109193652A (en) * | 2018-10-30 | 2019-01-11 | 贵州电网有限责任公司 | A kind of distribution network failure self-healing system containing distributed generation resource based on Situation Awareness |
| CN110224383A (en) * | 2019-06-28 | 2019-09-10 | 深圳供电局有限公司 | A kind of distribution network failure self-healing and line loss cooperative optimization method and system |
| CN111082401A (en) * | 2019-11-15 | 2020-04-28 | 国网河南省电力公司郑州供电公司 | Self-learning mechanism-based power distribution network fault recovery method |
| CN111751655A (en) * | 2020-05-29 | 2020-10-09 | 深圳供电局有限公司 | Fault self-healing method and device for distribution line, computer equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6867958B2 (en) * | 2000-12-27 | 2005-03-15 | Abb Technology Ag | Loop restoration scheme for distribution feeders |
-
2020
- 2020-11-04 CN CN202011214337.4A patent/CN112467686B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018025968A (en) * | 2016-08-10 | 2018-02-15 | 日本電信電話株式会社 | Restoration control system and method |
| CN108123439A (en) * | 2018-01-03 | 2018-06-05 | 燕山大学 | A kind of repairing of active distribution network multiple faults and restoration methods based on multi-agent system |
| CN109193652A (en) * | 2018-10-30 | 2019-01-11 | 贵州电网有限责任公司 | A kind of distribution network failure self-healing system containing distributed generation resource based on Situation Awareness |
| CN110224383A (en) * | 2019-06-28 | 2019-09-10 | 深圳供电局有限公司 | A kind of distribution network failure self-healing and line loss cooperative optimization method and system |
| CN111082401A (en) * | 2019-11-15 | 2020-04-28 | 国网河南省电力公司郑州供电公司 | Self-learning mechanism-based power distribution network fault recovery method |
| CN111751655A (en) * | 2020-05-29 | 2020-10-09 | 深圳供电局有限公司 | Fault self-healing method and device for distribution line, computer equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 考虑光伏波动性与负荷时变性的ADN多故障修复策略;孙秀飞等;分布式能源(第02期);第30-39页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112467686A (en) | 2021-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110943453B (en) | Power system fault sample generation and model construction method facing transfer learning | |
| US5708590A (en) | Method and apparatus for real time recursive parameter energy management system | |
| CN114006413B (en) | Power system transient stability control method and system based on graph neural network | |
| CN112467686B (en) | Power distribution network fault recovery method, device, control equipment and storage medium | |
| JP2022042469A (en) | Power generation prediction and efficiency diagnosis system for photovoltaic power generation facility using frbfnn model | |
| CN113759868A (en) | Industrial Ethernet fault prediction method based on neural network | |
| CN106771801B (en) | Online monitoring device for capacitor bank and application method of online monitoring device | |
| Lutska et al. | Forecasting the efficiency of the control system of the technological object on the basis of neural networks | |
| CN114152839A (en) | Power distribution network fault positioning method and system under distributed power grid-connected environment | |
| CN115693916B (en) | Intelligent online monitoring method and system for direct-current power supply of transformer substation | |
| CN116794510A (en) | Fault prediction method, device, computer equipment and storage medium | |
| CN113204280B (en) | Method, system, equipment and medium for diagnosing power failure | |
| CN112488224A (en) | Model training method and device, computer equipment and storage medium | |
| Weng et al. | An optimal control-based cascading failure risk assessment method | |
| Amiri et al. | A novel optimal PMU location method in smart grids | |
| CN110232482B (en) | Equipment management method and device based on neural network | |
| CN117192314B (en) | Insulation detection method and device based on insulation detection circuit and computer equipment | |
| CN117791597A (en) | Power distribution network fault self-healing method and system based on machine learning | |
| CN117722306A (en) | On-line monitoring method and system for water turbine speed regulator | |
| CN112600221A (en) | Reactive compensation device configuration method, device, equipment and storage medium | |
| CN117787915A (en) | Digital twin intelligent brain construction method for power distribution network | |
| CN116241993A (en) | Environment control method, system, device, electronic equipment and storage medium | |
| CN116260144A (en) | Power grid fragile line identification method, system, storage medium and computing device | |
| CN117272816A (en) | Method and device for predicting temperature difference of upper guide bush of water turbine | |
| CN114336612A (en) | Power loss calculation method for power station and fault equipment thereof and related application equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |